deploy

A deployment-phase workflow that prepares and, after approval, carries out a software release. It records the target, changed services, migrations, feature flags, rollback plan, affected systems, and verification details.

In plain words
What is it for?
Use it after testing has passed to prepare a deployment proposal, request human approval, execute the approved release under supervision, and record the result in deployment artifacts or configured tickets.
Why use it?
It makes release risks and recovery steps explicit before code reaches an environment. It also ensures deployment is approved by a person rather than happening automatically.

Skill for Claude CodeCodex

Part of the sdlc-plugin plugin — 19 skills, 16 commands, 10 agents, 4 hooks shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/lantisprime/claude-sdlc/deploy
Any agent
npx skills add lantisprime/claude-sdlc --skill deploy
Clone the repo
git clone --depth 1 https://github.com/lantisprime/claude-sdlc

Made for: Claude Code, Codex.

Or install sdlc-plugin, the plugin that ships this one along with the rest of its 19 skills, 16 commands, 10 agents, 4 hooks.

Per session 91 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 773 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00091 $0.00773
Opus 5 $0.00046 $0.00387
Sonnet 5 $0.00018 $0.00155
Haiku 4.5 $0.00009 $0.00077

Measured 2d ago against content hash db7a89ff63cc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

deploy scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/deploy/SKILL.md · 87 lines

How it starts

The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Deploy (Phase 6)

Release code that has passed Test — under explicit human approval.

Prerequisite

.claude/sdlc/gates/test-<task-slug>.md must exist and be signed. Without it, this skill refuses to run.

Step 1 — Propose the deployment

Produce a deployment proposal, not a deployment. Include:

  • Target environment (dev / staging / production)
  • Commit SHA / tag / image reference
  • Changed files and services
  • DB migrations? (yes/no, reversible?)
  • Feature flags? (default state)
  • Rollback plan
  • Blast radius (users/services affected)
  • Deployment window / maintenance window if applicable

Step 2 — Human approval gate

This is non-negotiable. Deployments never auto-execute. Ask the human to confirm the proposal. Record the approval at .claude/sdlc/gates/deploy-<task-slug>.md with timestamp and approver.

Step 3 — Execute (under human supervision)

Execute the deployment via the configured pipeline/runbook. Stream output. Halt on the first error and ask for direction — do not retry automatically.

Step 4 — Record

Update the ticket(s) with:

  • Commit SHA, environment, timestamp, approver
  • Link to pipeline run / deployment log
  • Post-deploy verification results (smoke tests)

Detection:

  • GitHub/GitLab Issues via .claude/sdlc/env.json → comment on the issue
  • Jira/Linear via MCP → transition to "Deployed" (or equivalent) with a comment
  • None configured → write .claude/sdlc/deployments/<YYYY-MM-DD>-<task-slug>.md (or .json if configured)

Step 5 — Post-deploy verification

Run the smoke-test subset of the test suite against the deployed environment. If any fail:

  • Execute the rollback from Step 1
  • Log a defect (see test skill)
  • Do not mark the deployment complete

What this skill must NEVER do

  • Auto-execute a deployment.
  • Skip the rollback plan.
  • Close a ticket before post-deploy verification passes.
  • Deploy code that does not have a signed test gate.

References

  • templates/deployment.md
  • templates/gate.md
  • docs/SDLC.md Deploy

Read the full file on GitHub · 87 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 87 lines · 91 tokens per session scan A db7a89ff63cc

Subscribe to this mod's changes

deploy is a skill published in the GitHub repository lantisprime/claude-sdlc (3 stars, last pushed 4mo ago), licensed MIT. It adds 91 tokens to every session and 773 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

release-check

Pre-release checklist for Forge — verify, version bump, changelog, docs, docker build, smoke test. Run before tagging v.

hoangsonww/Forge-Agentic-Coding-CLI · 30 tokens

landing

End-of-session ritual. Close finished beads, file new ones for remaining work, run verify, commit in logical groupings, push. Work is not complete until git push succeeds.

hoangsonww/Forge-Agentic-Coding-CLI · 38 tokens

live-bugbash

Safe-by-default Symphony for Trello bug bash. Uses fake Trello, fake Codex, and fake/local GitHub unless explicitly asked for a real live bug bash; the phrase "hardened host" enables run-scoped danger-full-access and Codex dangerous-bypass coverage. Covers SPEC.md conformance, installers, CLI options, workflow…

martin-francois/symphony-trello · 117 tokens

ship-it

Get the work done in this conversation all the way out — merged, and deployed to every environment the repo has, production included. Grounds itself first in the branches written during this session, the open and closed PRs behind them, and the environments and CI/CD workflows the repo actually defines, then lands and…

adamlinscott/claude-skills · 134 tokens

ship-it-now

Same skill as /ship-it, with the ride-along confirmation gate turned off.

adamlinscott/claude-skills · 135 tokens

tendril-release

Updates the used Ivy NuGet packages to the latest stable versions in a branch from development, builds/verifies, merges into development, creates a PR into main, merges it, merges main back to development, and triggers the GitHub Actions release workflow.

Ivy-Interactive/Ivy-Tendril · 55 tokens